- Más nuevo
- Más votos
- Más comentarios
I have gone through the bullet points on the article. Everything looks to be configured properly but in regards to this bulletpoint:
- Monitor the CPU and memory metrics of the service. For example, high CPU can make your application unresponsive and result in a 502 error.
What is considered high CPU? It appears that my CPU utilization starts off at very close to 100% before crashing. Could this be the issue and how can it be resolved?
Thanks. I increased the CPU/Memory to 4 vCPU/8 GB (previously 2 vCPU/4 GB) but the CPU utilization graph still looks the same and am having the same issue. I originally used 2 vCPU/4 GB because the documentation from Appsmith (the image I am trying to launch in my task) (https://docs.appsmith.com/getting-started/setup/installation-guides/aws-ecs-on-fargate) specified this in the deployment instructions.
Hmm..It seems its not resource issue. So my additional question is as follows.
· Can't you find containers launch successfuly from container logs? · Where 172.31.33.49 comes from? Tasks IP?Doesn't it change when task recreated?When I wrote cdk code about fargate I used
service.attachToApplicationTargetGroup
to specify target so specifying ip address is not needed.// Target Group const targetGroup = new elbv2.ApplicationTargetGroup(this, 'TargetGroup', { vpc, port: 80, protocol: elbv2.ApplicationProtocol.HTTP, targetType: elbv2.TargetType.IP, healthCheck: { path: '/', interval: Duration.seconds(60), healthyHttpCodes: '200' }, }) listener.addTargetGroups('TargetGroup', { targetGroups: [targetGroup], }) const service = new ecs.Ec2Service(this, 'Service', { cluster, taskDefinition, securityGroups: [securityGroupAPP], }) service.attachToApplicationTargetGroup(targetGroup) }
Hi This document will help you. https://repost.aws/knowledge-center/ecs-fargate-health-check-failures
Contenido relevante
- OFICIAL DE AWSActualizada hace un año
Can you confirm cpu/mem requirements of containers?Increasing CPU limit of task if it is possible.